Path planning methods of mobile robot based on new neural network

For the problem of traditional artificial neural network in robot path planning that unable to adapt to dynamic environment, this paper presents a new neural networks robot path planning algorithm. The arrangement of the neurons coincides with the discretized representation of configuration space. The target neuron has the maximal positive neural activity, which is damply promulgated to the whole state space via local lateral connections of neurons, robot was attracted to the target through the neural activity propagation, while the obstacles put away the robot to avoid collision by making themselves stay at the valley of the activity landscape. Simulation demonstrated that the generated path was continuous, smooth, and optimal, can respond quickly to the fast changing environment.